On Wednesday 31 August 2005 18:25, justin wrote: > Kern Sibbald wrote: > >On Wednesday 31 August 2005 10:33, justin wrote: > >>Hi, > >> > >><snip> > >> > >> > >>And that's it! The same thing happens with commands such as "label > >>barcodes". I'm running on Solaris 9 with a tape library. mtx and mt > >>commands work fine even if bacula is running - I don't think it's > >>anything to do with device contention. I attach what I think are the > >>relevant parts of the dir and sd conf files. Anyone know what's going > >> on? > > > >You will need to run it under the debugger as described in the Kaboom > > chapter of the manual. > > OK - done that. Some interesting results... > > First of all, I don't think there's anything really useful here - the > system *appears* to work OK now because of some of the things I did, but > it's nothing earth-shaking. > > I took Kern's advice and ran an instance of the SD under GDB with the > command line: > > run -s -f -d 99 -c /opt/bacula/etc/bacula-sd.conf > > > First off, I noticed that the SD was trying to access both the drives in > the changer but because there were no tapes in there at the time it > appeared to hang - waiting on I/O, I assume. Why this happened, I can't > say (but see below). I stopped this run, used mtx to load tapes into > the drives and tried again. This time the SD didn't hang on straight > queries, but crashed (with SIGPIPE) when I tried to do anything such as > get the status. > > I *did* notice the comments in the docs about compiling with the > --enable-thread-safe-client option, which I hadn't done before. (Did > this cause the problem?)
No, that is only for MySQL and is not critical. > Anyway, I reconfigured/compiled with this > option, scrubbed the database and tried again. Now it works - so far. > Kern, is the --enable-thread-safe-client *that* important? If so, I > think more emphasis on its importance in the documentation would be useful. > > Sorry guys, I know I changed a lot of variables at the same time which > is not too helpful. If anyone has ideas of how to track down possible > bugs I'll be happy to reconfigure and try to duplicate what you expect > to see. > > Now I appear to have hardware problems with the tape drives (I/O > errors). I'm investigating and I'll let you know if it appears to be > anything to to do with bacula. > > Regards, > > Justin. > > > ------------------------------------------------------- > SF.Net email is Sponsored by the Better Software Conference & EXPO > September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices > Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA > Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf > _______________________________________________ > Bacula-users mailing list > Bacula-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/bacula-users -- Best regards, Kern ("> /\ V_V ------------------------------------------------------- SF.Net email is Sponsored by the Better Software Conference & EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
